A support vector machine model to estimate phytoplankton size classes

  • YAO Lin-jie ,
  • CAO Wen-xi ,
  • WANG Gui-fen ,
  • XU Zhan-tang ,
  • HU Shui-bo ,
  • ZHOU Wen ,
  • LI Cai
Expand
  • 1.State Key Laboratory of Tropical Oceanography (South China Sea Institute of Oceanology, Chinese Academy of Sciences), Guangzhou 510301, China;
    2. University of Chinese academy of sciences, Beijing 100049, China;

Received date: 2014-10-02

  Online published: 2015-08-21

Abstract

In this study, a support vector machine (SVM) model was introduced to retrieve phytoplankton size classes (PSCs), from phytoplankton absorption spectra and total chlorophyll-a concentration.The performance of this model was validated with the South China Sea andNASA bio-Optical Marine Algorithm Dataset(NOMAD) datasets. The results of the model, which used phytoplankton absorption spectra as the only input parameter, showed that the absolute percentage differences (APD) were 46.1% (pico), 61.6% (nano) and 36.3% (micro) for the South China Sea dataset, and were 36.3% (pico), 44.6% (nano) ,44.3% (micro) for NOMAD dataset; It also showed that the determination coefficents (R2) were 0.604 (pico), 0.423 (nano) and 0.491 (micro) for the South China Sea dataset, and were 0.460 (pico), 0.702 (nano) and 0.829 (micro) for the NOMAD dataset. The SVM model that used both absorption spectra and chlorophyll-a concentration of phytoplankton showed that the APD were 19.2%, 31.9%, 31.6% and determination coefficents of phytoplankton size classes (pico, nano, and micro) were 0.837, 0.805, 0.600 for the South China Sea data-set. Using the same method, the result of SVM model showed that the APD were 35.3%, 35.4%, 38.2% and determination coefficents of phytoplankton size classes (pico, nano, and micro) were 0.713, 0.758, 0.810 for the NOMAD data set. The SVM model trained using phytoplankton absorption spectra and total chlorophyll-a concentration performed more effectively than that using only phytoplankton absorption spectra. The performance of the SVM model was shown to be satisfactory, and the model opens the way to an appliaction to estimate PSCs by using hyperspectral measurements.

Cite this article

YAO Lin-jie , CAO Wen-xi , WANG Gui-fen , XU Zhan-tang , HU Shui-bo , ZHOU Wen , LI Cai . A support vector machine model to estimate phytoplankton size classes[J]. Journal of Tropical Oceanography, 2015 , 34(4) : 37 -47 . DOI: 10.11978/j.issn.1009-5470.2015.04.005

References

1 梁少君, 曹文熙, 王桂芬, 等. 2010. 基于浮游植物吸收光谱提取粒径参数[J]. 热带海洋学报, 29(2): 59-64.
2 王桂芬, 曹文熙, 许大志, 等. 2005. 南海北部水体浮游植物比吸收系数的变化[J]. 热带海洋学报, 24(5): 1-10.
3 王桂芬, 曹文熙, 许大志, 等. 2007. 南海北部藻类粒级结构及色素成分对浮游植物吸收系数的影响[J]. 海洋学报(中文版), 29 (1): 38-48.
4 王桂芬, 曹文熙, 周雯, 等. 2010. 基于南海北部海区浮游植物吸收光谱斜率变化的粒级结构反演[J]. 热带海洋学报, 29(2): 25-32.
5 周雯, 曹文熙, 李彩, 等. 2010. 细胞结构对浮游植物光学特性的影响[J]. 热带海洋学报, 29(2): 33-38.
6 BOUMAN H, PLATT T, SATHYENDRANATH S, et al. 2005. Dependence of lightsaturated photosynthesis on temperature and community structure[J]. Deep-Sea Research Ⅰ, 52(7): 1284-1299.
7 BREWIN R J W, SATHYENDRANATH S, HIRATA T, et al. 2010. A three-component model of phytoplankton size class for the Atlantic Ocean[J]. Ecological Modelling, 221(11): 1472-1483.
8 BREWIN R J W, HARDMAN-MOUNTFORD N J, LAVENDER S J, et al. 2011. An intercomparison of bio-optical techniques for detecting dominant phytoplankton size class from satellite remote sensing[J]. Remote Sensing of Environment, 115(2): 325-339.
9 BRICAUD A, MEJIA C, BIONDEAU-PATISSIER D, et al. 2007. Retrieval of pigment concentrations and size structure of algal populations from their absorption spectra using multilayered perceptrons[J]. Applied Optics, 46(8): 1251-1260.
10 CADDY J, REFK R, DO-CHI T. 1995. Productivity estimates for the Mediterranean: evidence of accelerating ecological change[J]. Ocean & Coastal Management, 26(1): 1-18.
11 CAMPBELL J W, ESAIAS W E. 1983. Basis for spectral curvature algorithms in remote sensing of chlorophyll[J]. Applied Optics, 22(7): 1084-1093.
12 CAO W X, YANG Y Z, XU X Q, et al. 2003. Regional patterns of particulate spectral absorption in the Pearl River estuary[J]. Chinese Science Bulletin, 48(21): 2344-2351.
13 CHANG C C, LIN C J. 2001. Training nu-support vector classifiers: Theory and algorithms[J]. Neural Computation, 13(9): 2119-2147.
14 CHISHOLM S W. 1992. Phytoplankton size[M] // FALKOWSKI P G, WOODHEAD A D. Primary Productivity And Biogeochemical Cycles in the Sea. New York: Plenum: 213-237.
15 DEVRED E, SATHYENDRANATH S, STUART V, et al. 2006. A two-component model of phytoplankton absorption in the open ocean: Theory and applications[J]. Journal of Geophysical Research-Oceans, 111(C3): 11.
16 HIRATA T, AIKEN J, HARDMAN-MOUNTFORD N, et al. 2008. An absorption model to determine phytoplankton size classes from satellite ocean colour[J]. Remote Sensing of Environment, 112(6): 3153-3159.
17 HOGE F E, SWIFT R N. 1986. Chlorophyll pigment concentration using spectral curvature algorithms - An evaluation of present and proposed satellite ocean color sensor bands[J]. Applied Optics, 25(20): 3677-3682.
18 HOUSE J I, COLIN PRENTICE I, LE QUÉRÉ C. 2002. Maximum impacts of future reforestation or deforestation on atmospheric CO 2 [J]. Global Change Biology, 8(11): 1047-1052.
19 HUANG B-Q, LIN X-J, LIU Y, et al. 2002. Ecological study of picoplankton in northern South China Sea[J]. Chinese Journal of Oceanology and Limnology, 20(Special Issue): 22-32.
20 LEE Z, CARDER K L. 2000. Band-ratio or spectral-curvature algorithms for satellite remote sensing?[J]. Applied Optics, 39(24): 4377-4380.
21 LI Z, LI L, SONG K, et al. 2013. Estimation of phytoplankton size fractions based on spectral features of remote sensing ocean color data[J]. Journal of Geophysical Research-Oceans, 118(3): 1445-1458.
22 LIN J, CAO W, ZHOU W, et al. 2013. A bio-optical inversion model to retrieve absorption contributions and phytoplankton size structure from total minus water spectral absorption using genetic algorithm[J]. Chinese Journal of Oceanology and Limnology, 31(5): 970-978.
23 LUBAC B, LOISEL H. 2007. Variability and classification of remote sensing reflectance spectra in the eastern English Channel and southern North Sea[J]. Remote Sensing of Environment, 110(1): 45-58.
24 MUELLER J L, FARGION G S, MCCLAIN C R, et al. 2003. Ocean optics protocols for satellite ocean color sensor validation, revision 4, volume Ⅳ: Inherent optical properties: Instruments, characterizations, field measurements and data analysis protocols[J]. NASA Tech. Memo: 01674-01670.
25 NAIR A, SATHYENDRANATH S, PLATT T, et al. 2008. Remote sensing of phytoplankton functional types[J]. Remote Sensing of Environment, 112(8): 3366-3375.
26 ORGANELLI E, BRICAUD A, ANTOINE D, et al. 2013. Multivariate approach for the retrieval of phytoplankton size structure from measured light absorption spectra in the Mediterranean Sea (BOUSSOLE site)[J]. Applied Optics, 52(11): 2257-2273.
27 PARSONS T, LALLI C. 2002. Jellyfish population explosions: revisiting a hypothesis of possible causes[J]. La Mer, 40: 111-121.
28 PLATT T, DENMAN K. 1977. Organization In Pelagic Ecosystem[J]. Helgolander Wissenschaftliche Meeresuntersuchungen, 30(1-4): 575-581.
29 PLATT T, BOUMAN H, DEVRED E, et al. 2005. Physical forcing and phytoplankton distributions[J]. Scientia Marina, 69(S1): 55-73.
30 RAITSOS D E, LAVENDER S J, MARAVELIAS C D, et al. 2008. Identifying four phytoplankton functional types from space: An ecological approach[J]. Limnology and Oceanography, 53(2): 605-613.
31 SATHYENDRANATH S, COTA G, STUART V, et al. 2001. Remote sensing of phytoplankton pigments: a comparison of empirical and theoretical approaches[J]. International Journal of Remote Sensing, 22(2-3): 249-273.
32 SIEBURTH J, SMETACEK V, LENZ J. 1978. Pelagic ecosystem structure: heterotrophic compartments of the plankton and their relationship to plankton size fractions[J]. LimnolOceanogr, 23(6): 1256-1263.
33 UITZ J, CLAUSTRE H, MOREL A, et al. 2006. Vertical distribution of phytoplankton communities in open ocean: An assessment based on surface chlorophyll[J]. Journal of Geophysical Research-Oceans, 111(C8): 23.
34 VAPNIK V N, CHERVONE A Y. 1971. Uniform convergence of relative frequencies of events to their probabilities[J]. Theory of Probility and Its Applications,16(2): 264-280.
35 VAPNIK V N. 1995. The Nature of Statistical Learning Theory[M]. New York: Springer-Verlag New York: 344.
36 VAPNIK V N, VAPNIK V. 1998. Statistical learning theory[M]. City: Wiley New York.
37 VIDUSSI F, CLAUSTRE H, BUSTILLOSGUZMAN J, et al. 1996. Determination of chlorophylls and carotenoids of marine phytoplankton: Separation of chlorophyll a from divinyl-chlorophyll a and zeaxanthin from lutein[J]. Journal of plankton research, 18(12): 2377-2382.
38 VIDUSSI F, CLAUSTRE H, MANCA B B, et al. 2001. Phytoplankton pigment distribution in relation to upper thermocline circulation in the eastern Mediterranean Sea during winter[J]. Journal of Geophysical Research-Oceans, 106(C9): 19939-19956.
39 WANG G, CAO W, YANG Y, et al. 2010. Variations in light absorption properties during a phytoplankton bloom in the Pearl River estuary[J]. Continental Shelf Research, 30(9): 1085-1094.
40 WANG G, CAO W, WANG G, et al. 2013. Phytoplankton size class derived from phytoplankton absorption and chlorophyll-a concentrations in the northern South China Sea[J]. Chinese Journal of Oceanology and Limnology, 31(4): 750-761.
41 WERDELL P J, BAILEY S W. 2005. An improved in-situ bio-optical data set for ocean color algorithm development and satellite data product validation[J]. Remote Sensing of Environment, 98(1): 122-140.
42 ZHAI H C, NING X R, TANG X X, et al. 2011. Phytoplankton pigment patterns and community composition in the northern South China Sea during winter[J]. Chinese Journal of Oceanology and Limnology, 29(2): 233-245.

Outlines

/